RFR jextract generates invalid generic signature for pointer to a function typedef

[Henry Jen]

Good catch of the bug. The webrev looks good.

In HeaderFile.java, it’s probably better  to unwrap type at the native level instead of JType. The define process is to figure out Java type for a native type after all.

--- a/src/jdk.jextract/share/classes/com/sun/tools/jextract/HeaderFile.java     Thu Jul 05 08:43:46 2018 -0700
+++ b/src/jdk.jextract/share/classes/com/sun/tools/jextract/HeaderFile.java     Mon Jul 09 08:50:04 2018 -0700
@@ -253,7 +253,7 @@
                 jt = doRecord(t);
                 break;
             case Pointer:
-                Type pointee = t.getPointeeType();
+                Type pointee = t.getPointeeType().canonicalType();
                 jt2 = (JType2) globalLookup(pointee);
                 jt = jt2.getDelegate();
                 if (jt instanceof JType.Function) {

Cheers,
Henry



> On Jul 9, 2018, at 8:36 AM, Sundararajan Athijegannathan <sundararajan.athijegannathan at oracle.com> wrote:
> 
> Please review: http://cr.openjdk.java.net/~sundar/pointer_to_function_typedef/webrev.00/
> 
> Thanks,
> -Sundar